Edward N. Wentworth 117 



descendants of four pairs medium in apparent vigor. Since both show 

 the same qualitative results they will be combined for the comparison 

 with Tables I and II. The means of the separate generations follow : 



i^. Mean 78-875 



The mean of the entire lot is 72 06, and the standard deviation 

 is 37-1618. This gives the largest coefficient of all, 51-27, and from 

 the standpoint of the coefficient would offer evidence of a 1 : 2 : 1 ratio. 

 That this is merely apparent will be shown a little later in the paper. 

 Summarized we get the following table. 



standard Coefficient 



Mean Deviation of Variation 



Low line ... 29-5 7-3 28-17 



Medium line ... 72-06 37-2 51-57 



High line ... 135 -80 18-3 13-47 



A segregation of fecundity factors is clearly evidenced and the 

 supposed weaknesses from inbreeding are shown up in their true light 

 as the mere segregation of factors for lower vigor. 



In order to confirm the idea of the relative separateness of these 

 fecundity factors a reciprocal cross was arranged between flies from 

 the hatch of 142 in Line III, pair A, Ft, and flies from the hatch of 

 30 in Line IV, pair C, F^. Table V shows the detailed results of this 

 work, the confirmation being absolute. The interesting point brought 

 out is the fact that the male, whether he come from high or low lines, 

 apparently in nowise influences the eggs laid by the female with whom 

 he pairs, though marked differences, apparently due to segregation, may 

 occur among his female descendants. 



Further evidence on the nature of such segregation was obtained 

 by breeding all the pairs available in one hatch from an apparently 

 intermediate pair. The pair selected was from Line IV, pair B, F^, 

 Hatch 82. Thirty-nine males and forty-three females appeared and 



